The invention disclosed herein relates generally to atomic standards and a method of providing a frequency and/or time based on an atomic standard, and more particularly to atomic frequency and/or time standards and methods employing a beam tube.
Current commercially-available atomic frequency or time standards typically employ a Cesium beam tube to control the frequency of a crystal-controlled oscillator. Such commercially-available units have a classical Cesium beam tube configuration. The construction and operation of classical Cesium beam tube atomic frequency standards is described, for example, in U.S. Pat. No. 2,972,115 (Zacharias et al.) and in Molecular Beams, by Norman F. Ramsey, Oxford University Press, 1956, pp 283-285. The development of atomic frequency standards is described in NBS Monograph, Time and Frequency REB Theory and Fundamentals, "A Historical Review of Atomic Frequency Standards", Chapter 4, pages 85-109, by Roger E. Beehler, 1974.
FIG. 1 of the drawings herein shows schematically a classical Cesium ("Cs") beam frequency standard 10 which includes a Cs source (oven) 12, a state preparation or selection region ("A" region) 14, a oscillator 17, a microwave cavity 16, a state selection region ("B" region) 18, and a detector 20. In the classical Cs beam standard, cesium atoms emitted from oven 12 are directed into the gap 24 of the "A" magnet 22. Due to the large gradient of the "A" magnet 22, the cesium atoms in the F=3 and F=4 states are deflected in different directions. In the particular arrangement shown in FIG. 1, the "A" magnet 22 deflects the cesium atoms in the F=3 state to and through the microwave cavity 16 to the "B" magnet 28 The magnetic field in cavity 16 created by microwave oscillations -5 causes some of the Cs atoms in cavity 16 undergo a transition to the F=4 state. The atoms that have undergone a transition from F=3 to F=4 in the cavity 16 are deflected towards the atomic detector 20 by the "B" magnet 28.
The output 26 of detector 20 is coupled to oscillator 17. The detection signal on output 26 of detector 20 varies in accordance with the quantity of Cs atoms that have undergone a transition in the cavity 16, and functions as an error signal for controlling the frequency of oscillator 17.
A Cs beam tube atomic standard employing optical pumping for state preparation and optical pumping for detection has been the subject of investigation for several years. Such optically pumped Cs ("OPCS") beam tube standards employ optical state selection and state detection in the "A" and "B" regions, as well as a microwave cavity and oscillator which may be quite similar to the ones used in the classical Cs beam tube standard described above. In the "A" state preparation region, optical pumping creates a hyperfine population difference. In the "B" region, the resonance phenomenon is detected through fluorescent light variations as a function of the microwave frequency. This fluorescent light variation, which is a direct measure of the number of Cs atoms which have undergone a transition, is detected and used to control the frequency of the microwave oscillations. A number of different pumping schemes are possible which may prepare cesium atoms in either .sup.2 S.sub.1/2, F=3 or F=4 hyperfine states (see the energy level diagram of FIG. 2). Such schemes are well known and described in the literature. The schemes are summarized in Table 1.
TABLE 1 ______________________________________ No. of Pump Laser Probe Laser Lasers Transition Transition ______________________________________ 1 .sup. (F .fwdarw. F') (F .fwdarw. F') 4 .fwdarw. 5* 4 .fwdarw. 5 4 .fwdarw. 4 4 .fwdarw. 4 4 .fwdarw. 3 4 .fwdarw. 3 3 .fwdarw. 4 3 .fwdarw. 4 3 .fwdarw. 3 3 .fwdarw. 3 3 .fwdarw. 2 3 .fwdarw. 2* 2 4 .fwdarw. 4 4 .fwdarw. 5* 4 .fwdarw. 3 4 .fwdarw. 5* 3 .fwdarw. 3 3 .fwdarw. 2* 3 .fwdarw. 4 3 .fwdarw. 2* 3 3 .fwdarw. 3 and 4 .fwdarw. 4.pi. 3 .fwdarw. 2* 3 .fwdarw. 4 and 4 .fwdarw. 4.pi. 3 .fwdarw. 2* 4 .fwdarw. 3 and 3 .fwdarw. 3.pi. 4 .fwdarw. 5* 4 .fwdarw. 4 and 3 .fwdarw. 3.pi. 4 .fwdarw. 5* ______________________________________ *Cycling transition
As far as applicant is aware, such OPCS beam tube standards are either of the laboratory-type or are in development, and are not commercially available. For example, Frequency Electronics, Inc., the assignee of this application, has been involved in development of a commercial OPCS beam tube standard since 1983, and has produced several prototypes for study and evaluation, which are described in 41st Ann. SFC 59 and 39th Ann. SFC 18, both cited below.
FIG. 3 is a schematic diagram of an OPCS beam tube standard 40. Similar to the classical Cs beam standard 10, the OPCS beam tube standard 40 uses an oscillator 17a and a microwave cavity 16; however, in the "A" state preparation region 14, and in the "B" detection region 18, laser sources 44 and 46 are employed for optical pumping rather than "A" and "B" magnets for state selection. In the OPCS beam tube standard depicted in FIG. 3, different laser sources 44 and 46 are employed. Cs source -2 is axially aligned with the central axis 25 of cavity 16 since laser source 44, unlike a magnet, does not deflect Cs atoms having different states in different directions. Rather, the Cs atoms from source 12 are optically pumped by laser 44 into a hyperfine state in the "A" region 14.
The Cs atoms then enter microwave cavity 16. The magnetic field in cavity 16 created by microwave oscillations 15acauses some of the Cs atoms passing through cavity 16 to undergo a transition as a result of resonance phenomenon.
After leaving cavity 16, Cs atoms enter the "B" region 18, where the resonance phenomenon in cavity 16 is detected through fluorescent light variations as a function of the microwave frequency. This is accomplished by optically pumping Cs atoms back to the hyperfine state, which generates photons and the fluorescence described above.
Fluorescence detector 48 is tuned to detect photon emitted by the Cs atoms as they change their hyperfine state. Fluorescence detector 48 has an output 50 coupled to oscillator 17a. The detection signal on detector output 50 varies in accordance with the detected fluorescence, and functions as an error signal for controlling the frequency of oscillator 17a.
A discussion of optically-pumped beam tube atomic standards may be found in the following published articles: G. Singh, P. Dilavore, C.O. Alley, IEEE J. Quant. Electr. QE-7, p. 196, 1971; M. Arditi, J-L Picque, J. Phys. 41, pp 6-379, 1980; L. L. Lewis, M. Feldman, 35th Ann. SFC 612, 1981; A. Derbyshire, R. E. Drullinger, M. Feldman, D. J. Glaze, D. Hillard, D. A. Howe, L. L. Lewis, J. H. Shirly, I. Pascaru, D. Stanciulescu, 39th Ann. SFC 18, 1985; G. Avila, E. Clerq, M. Labachelerie, P. Cerez, IEEE Trans. Instr. Meas. IM-34, p. 139, 1985; T. McClelland, I. Pascaru, J. Zacharski, N. H. Tran, M. Meirs, 41st Ann. SFC 59, 1987; C. Jacques and P. Tremblay, 42nd Ann. SFC, 1988, pp 505-509; and Frequency Standards and Metrology, A. DeMarchi, editor, Springer - Verlag Berlin, Hesdelberg, 1989, pp 110-136, 281-386, 391-394.
The OPCS frequency standard under development by Frequency Electronics, Inc. is described in 41st Ann. SFC 59 and 39th Ann. SFC 18, both cited above.
Optical pumping is expected to improve the frequency stability and accuracy of Cesium beam frequency standards. For example, as described in 1981 35th Ann. SFC 612, cited above, it was believed that optical pumping by laser diodes would increase stability in Cs beam tube standards because optical pumping in the "A" region causes Cs atoms to be converted to the desired hyperfine state rather than being rejected by the magnet, which would provide more atoms for the ultimate detection signal for a given beam intensity. This should provide an improvement in the signal to noise ratio of the standard. The number of levels involved in the optical pumping of Cs in the ground state is 16. With a single laser pumping in the "A" region, Cs atoms will be distributed among nine Zeeman sublevels, with a population distribution of approximately 13% per level. If two lasers are used for optical pumping in "A" region, it is theoretically possible to have a complete population inversion. This corresponds to a significant increase in the intensity of the output signal.
Use of optical pumping in the "B" region relaxes the stringent alignment otherwise necessary for a magnet in the "B" region. Fluorescence detection should have long term stability since all atoms in the beam would be sensed regardless of velocity and position. Also, detection by fluorescence on a state transition produces a large number of photons from each atom of the Cesium beam which should facilitate detection. These also should improve the signal to noise ratio. Optical pumping should also reduce the size of some systematic frequency offsets.
However, the use of optically-pumped state preparation and optically-pumped detection has some drawbacks which have prevented, so far as applicants are aware, full realization of the improvements expected. The intrinsic noise of the optical detector and the shot noise of the stray light are the main limitation factors in achieving the expected improvement in signal to noise ratio. The laser frequency noise is transformed into fluorescence photon noise via the state transition process, and represents another limiting factor in obtaining the expected improved signal to noise ratio. The quality factor ("Q") of the resonance line is less than that in classical Cesium tubes since the velocity selection on the low side of the Maxwellian distribution obtained with a magnetic field does not occur with optical pumping. Also, power and frequency fluctuations of currently available laser diodes degrade the performance of the Cesium beam tube.
Investigation of optically-pumped beam tube standards including studying these limitations and obtaining experimental results has been described in the literature for several years (see, for example, the documents cited above). The literature describes different approaches to diminish these limitations. However, to applicants' knowledge, no commercial optically-pumped beam tube standard is yet available, and to applicants' knowledge, no optically-pumped beam tube standard has achieved the performance eventually expected of optically-pumped beam tube standards.